Deepgram Pricing 2026: Nova-3 $0.26/hr Batch vs Streaming

Deepgram's pricing has a decision point that changes your bill more than any plan choice: pre-recorded (batch) vs streaming. On pay-as-you-go, Nova-3 costs $0.0043/min for pre-recorded audio (about $0.26/hour) and $0.0048/min for streaming at the current promotional rate ($0.0077/min regular, about $0.46/hour). Speaker diarization is included on pre-recorded audio and costs $0.0020/min extra on streaming. All figures were verified on Deepgram's pricing page on September 8, 2026.

This price difference isn't arbitrary. Real-time streaming requires dedicated infrastructure, instant processing, and WebSocket management. But here's the question most developers don't ask until after implementation: does your use case actually need real-time, or can you tolerate a few minutes of delay?

In this guide, we break down Deepgram's complete 2026 pricing across pre-recorded and streaming, explain the pay-as-you-go vs Growth plan differences, show what real files from our own production database would cost on each option, and explain when a simpler $6.00 flat rate makes more sense. If batch is acceptable and you'd prefer not to maintain an integration at all, BrassTranscripts' speech-to-text with automatic speaker labels turns a file upload into a finished transcript in roughly 2–5 minutes per hour of audio.

For comparing transcription pricing across all major services, see our comprehensive cost analysis.

Quick Navigation

• Deepgram Pricing Overview (2026)
• What Real Files Cost on Deepgram vs BrassTranscripts (Our Production Data)
• Real-Time vs Batch: When Does the Premium Make Sense?
• Per-Second Billing: Deepgram's Advantage
• Hidden Costs Most Developers Miss
• Deepgram vs BrassTranscripts: When Simplicity Wins
• Real-World Cost Scenarios
• Deepgram Free Tier & Volume Discounts
• When to Choose Deepgram vs Alternatives
• Frequently Asked Questions
• AI Prompt: Deepgram Pricing Calculator
• Final Verdict: Deepgram vs BrassTranscripts
• Pricing Disclaimer

Deepgram Pricing Overview (2026)

Deepgram charges $0.0043 per minute for pre-recorded Nova-3 transcription on pay-as-you-go and $0.0036 on the Growth plan, with streaming priced separately and diarization included only on pre-recorded audio. Here is the structure from Deepgram's pricing page, verified September 8, 2026:

Nova-3 Pricing (Current Model)

Plan Differences

• Pay-As-You-Go: No minimum commitment, $200 in free credit to start
• Growth Plan: Annual commitment starting at $4,000/year, pre-recorded rate drops 16% to $0.0036/min

Last verified: September 8, 2026 from Deepgram Pricing. The streaming "current price" is labeled as promotional on Deepgram's page; the regular price is shown alongside it.

Cost Calculation at 1,000 Hours/Month (Pre-Recorded)

Pay-As-You-Go: 1,000 hours × 60 min × $0.0043 = $258/month
Growth Plan:   1,000 hours × 60 min × $0.0036 = $216/month
Savings with Growth: $42/month (16% less)

The Growth plan saves money but requires an annual commitment. Pay-as-you-go offers flexibility without lock-in.

What Real Files Cost on Deepgram vs BrassTranscripts (Our Production Data)

On real customer files, Deepgram's pre-recorded API costs between six cents and fifty cents per file, while BrassTranscripts charges $2.50 or $6.00 flat; the gap is the price of not writing or maintaining an integration.

Two things stand out. First, at these file lengths the streaming premium is small in dollars, so the real reason to avoid streaming is engineering complexity, not price. Second, the flat-rate gap is between $2.44 and $5.88 per file. For a team transcribing a handful of meetings a month that gap is less than the cost of an hour of developer time; for a product processing thousands of files a day it is decisive. The scenarios below work through both ends.

Real-Time vs Batch: When Does the Premium Make Sense?

Real-time streaming costs 12% more than pre-recorded at Deepgram's current promotional rate and 79% more at the regular rate, before the $0.0020/min streaming diarization add-on; the premium is only worth paying when instant results are functionally required.

Use Cases Where Real-Time WINS (Premium Justified)

1. Live Captioning

• Webinars, conferences, virtual events
• Users see captions as speech happens
• Latency requirements: < 3 seconds
• Real-time required: YES

2. Voice Assistants / Smart Devices

• Conversational AI applications
• User expects immediate response
• Real-time required: YES (this is what Deepgram's Flux model targets)

3. Live Customer Support

• Call center agent assist (real-time suggestions)
• Compliance monitoring during live calls
• Instant sentiment detection
• Real-time required: YES

4. Accessibility Compliance

• Live captioning for deaf and hard-of-hearing attendees
• Legal requirement, not optional
• Real-time required: YES

Real-time premium justified: when instant results are functionally required or legally mandated.

Use Cases Where Batch WINS (Premium Not Justified)

1. Podcast Transcription

• Content already recorded
• Can wait minutes for results
• Users download transcripts after publishing
• Batch acceptable: YES

2. Meeting Notes

• Transcribe after the meeting concludes
• Team reviews notes hours or days later
• Batch acceptable: YES

3. Interview Transcription

• Research, journalism, legal discovery
• Transcripts reviewed days or weeks later
• Batch acceptable: YES

4. Video Content Transcription

• YouTube videos, courses, webinars
• Post-production workflow
• Batch acceptable: YES

5. Voicemail Transcription

• Async communication by nature
• Minutes of delay irrelevant
• Batch acceptable: YES

Batch wins: when you can tolerate a few minutes of processing delay, which describes most recorded-audio workloads.

Per-Second Billing: Deepgram's Advantage

Deepgram bills by the actual second rather than rounding up to the minute, which matters most for workloads made of many short clips.

Example: 61-second audio file

• Deepgram charges: 61 seconds = $0.00437 (at $0.0043/min pre-recorded PAYG)
• A competitor rounding to the minute: 2 minutes = $0.006 (at a $0.003/min rate)

Advantage: fairer billing for short audio clips.

Real-world impact for customer service calls:

1,000 calls/month, average 45 seconds each

Deepgram billing (pre-recorded):
1,000 × 45 seconds = 45,000 seconds = 750 minutes
750 min × $0.0043 = $3.23/month

Competitor (per-minute rounding):
1,000 × 1 minute (rounded up) = 1,000 minutes
1,000 min × $0.003 = $3.00/month

In this case, per-second billing doesn't save money (the competitor's lower rate wins). But it's mathematically fairer, and at Deepgram's rate the whole month costs less than a coffee either way.

Hidden Costs Most Developers Miss

The hidden cost of Deepgram is engineering time, not the per-minute rate: a production-ready streaming integration takes roughly 8-16 hours to build, and the Growth plan's annual minimum is sunk cost if usage drops.

1. Real-Time Infrastructure Requirements

Real-time streaming isn't just more expensive per minute; it requires different architecture:

WebSocket Management:

• Persistent connection handling
• Connection pooling for multiple streams
• Reconnection logic for dropped connections
• Development time: 4-8 hours

Audio Streaming:

• Chunk audio into small packets
• Handle backpressure and buffering
• Manage microphone permissions (browser)
• Development time: 6-12 hours

Real-World AWS Cost (for real-time streaming infrastructure):

API Gateway WebSocket: $1.00/million minutes connected
Lambda (processing): $0.20/million invocations
DynamoDB (connection state): $1.25/million writes
─────────────────────────────────────────────────────
Infrastructure: ~$2.45/month (low volume)

At 1,000 hours/month streaming, infrastructure adds roughly $8-15/month overhead.

2. The Streaming Diarization Add-On

Speaker diarization is included on pre-recorded audio but is a separate $0.0020/min line item on streaming. On the promotional streaming rate that raises the effective price from $0.0048 to $0.0068/min, a 42% increase, and it is easy to miss when estimating from the headline number.

3. The "Growth Plan" Consideration

Deepgram offers annual commitment discounts:

• Pay-As-You-Go: no commitment, $0.0043/min pre-recorded
• Growth Plan: from $4,000/year, $0.0036/min pre-recorded (16% savings)

The consideration: Growth plans require an annual commitment. If your usage drops or you switch providers, the remaining commitment is sunk cost. Spending the $4,000 minimum at the Growth rate takes about 1.1 million minutes, or roughly 1,500 hours of pre-recorded audio per month.

Better approach: start with pay-as-you-go. Only commit to annual once you've validated 6+ months of consistent usage.

4. Free Tier Limitations

Deepgram offers $200 in free credit, generous compared to competitors. But:

• Credit is one-time, not recurring monthly
• Once exhausted, you're immediately on the paid tier

$200 value:

Pay-As-You-Go (pre-recorded): $200 ÷ $0.0043/min = 46,512 minutes ≈ 775 hours
Streaming (promo rate):       $200 ÷ $0.0048/min = 41,667 minutes ≈ 694 hours

Excellent for testing, but plan your budget for post-credit costs.

Deepgram vs BrassTranscripts: When Simplicity Wins

Deepgram is cheaper per hour at every volume; BrassTranscripts wins only when the cost of building and maintaining an API integration outweighs a flat $2.50 or $6.00 per file, which is typically true below a few dozen files a month.

Where Deepgram Wins

1. Competitive API Pricing At $0.0043/min pay-as-you-go or $0.0036/min on Growth, Deepgram is among the cheapest managed speech-to-text APIs. For 10,000 hours/month with API capability:

10,000 hours × 60 min × $0.0036 = $2,160/month (Growth)
10,000 hours × 60 min × $0.0043 = $2,580/month (PAYG)

2. Real-Time Streaming Capability If you genuinely need real-time transcription (live captioning, voice agents), Deepgram's streaming API and Flux model are built for it.

3. Per-Second Billing Fairness For short audio clips, per-second billing ensures you're not overpaying for unused minutes.

4. Diarization Included on Pre-Recorded Audio Unlike some API vendors, Deepgram does not charge extra for speaker labels on batch jobs.

Where BrassTranscripts Wins

1. No API Required: Upload and Go Deepgram is API-only. You must handle:

• WebSocket connections (for real-time)
• Async job handling (for batch)
• Audio format validation
• Error handling and retries

BrassTranscripts: upload file, download transcript. Zero technical complexity.

2. No Account Needed Deepgram requires account creation, API key management, and payment setup. BrassTranscripts lets you upload immediately, preview the first 30 words, and pay only if you want the full transcript.

3. Predictable Pricing BrassTranscripts: $2.50 flat for files up to 15 minutes, $6.00 flat for longer files, speaker labels and four output formats included. No mental math about batch vs streaming, no add-ons, no surprise bills.

Cost Comparison: 200 Hours/Month (200 one-hour files)

At 200 files a month the API pays for its own integration within the first month. The picture flips at low volume: at 10 one-hour files a month the difference is $60 vs $2.58, about $57/month, which does not justify a build unless you have engineers idle.

Real-World Cost Scenarios

Across four realistic workloads, Deepgram costs between 3% and 0.04% of the BrassTranscripts price; the only scenario where the flat rate wins is a non-technical team with modest volume.

Scenario 1: Podcast Network (100 Episodes/Month)

Requirements:

• 100 episodes/month, average 45 minutes each
• Transcription for SEO and accessibility
• Published days after recording (no time pressure)
• Non-technical podcast producers

Deepgram Option (PAYG, pre-recorded):

Audio: 100 × 45 min = 4,500 minutes/month
Nova-3: 4,500 × $0.0043 = $19.35
Speaker ID: included
API development: not feasible without a developer
───────────────────────────────────────────────────
$19.35/month once someone builds the integration

BrassTranscripts Option:

Audio: 100 files, all over 15 minutes
Rate: $6.00 flat per file (includes speaker ID)
───────────────────────────────────────────────────
Total: $600/month
No technical skills required

Winner: depends on who is on the team. A network with one part-time developer saves about $580/month with Deepgram. A network of producers with no developer pays $600 and never thinks about it.

Scenario 2: Virtual Event Platform (Live Captioning)

Requirements:

• 500 hours/month live events
• Real-time captions required for accessibility
• Has a development team

Deepgram Option (PAYG, streaming):

Audio: 500 hours × 60 = 30,000 minutes
Nova-3 streaming: 30,000 × $0.0048 = $144.00 (promo; $231.00 at the regular $0.0077 rate)
Speaker ID add-on: 30,000 × $0.0020 = $60.00
Infrastructure: ~$12/month
───────────────────────────────────────────────────
Total: ~$216/month (promo) to ~$303/month (regular)

BrassTranscripts Option:

Real-time transcription: NOT AVAILABLE
BrassTranscripts is batch-only
───────────────────────────────────────────────────
Cannot use for live captioning

Winner: Deepgram. Real-time streaming is functionally required; BrassTranscripts can't serve this use case.

Scenario 3: Research University (Interview Archive)

Requirements:

• 300 one-hour interviews/month
• Transcripts for qualitative analysis
• No time sensitivity
• Mixed technical capability

Deepgram Option (PAYG, pre-recorded):

Audio: 300 hours × 60 = 18,000 minutes
Nova-3: 18,000 × $0.0043 = $77.40
Speaker ID: included
API development: $800-1,200 (one-time)
───────────────────────────────────────────────────
First month: $877.40-1,277.40
Ongoing: $77.40/month

BrassTranscripts Option:

Audio: 300 files over 15 minutes
Rate: $6.00 flat per file
───────────────────────────────────────────────────
Total: $1,800/month
No development required

Winner: Deepgram if the department has a developer or a research-computing group; the integration pays for itself in the first month. BrassTranscripts for a lab that needs transcripts this week and has no one to build anything.

Scenario 4: Mobile App (Voice Memos)

Requirements:

• 10,000 short voice memos/month
• Average 12 seconds each
• Users expect near-instant transcription
• Startup with an engineering team

Deepgram Option (PAYG, streaming):

Audio: 10,000 × 12 sec = 120,000 seconds = 2,000 minutes
Nova-3 streaming: 2,000 × $0.0048 = $9.60 (promo; $15.40 regular)
Infrastructure (WebSocket): ~$8/month
───────────────────────────────────────────────────
Total: ~$18-23/month
Per-second billing advantage for short clips

BrassTranscripts Option:

10,000 files × $2.50 short-file rate = $25,000/month
───────────────────────────────────────────────────
Not designed for this: per-file pricing and batch
processing don't fit thousands of 12-second clips

Winner: Deepgram, by three orders of magnitude. Per-file flat rates are built for humans uploading recordings, not for products transcribing thousands of clips.

Deepgram Free Tier & Volume Discounts

Deepgram's $200 signup credit covers about 775 hours of pre-recorded Nova-3 audio at pay-as-you-go rates, enough to run a serious evaluation before spending anything.

Free Tier ($200 Credit)

Deepgram provides $200 in free credit for new accounts:

• Pre-recorded: about 775 hours at $0.0043/min
• Streaming: about 694 hours at the $0.0048/min promotional rate

Best use: test accuracy on representative audio across batch and streaming modes before committing to paid usage.

Volume Discounts (Growth Plan)

Deepgram offers annual commitment discounts:

• Commitment: from $4,000/year
• Growth Plan rate: $0.0036/min pre-recorded (16% savings vs $0.0043/min PAYG); streaming rates are custom
• Requirements: annual commitment

Break-even analysis:

Pay-As-You-Go: $0.0043/min
Growth Plan:   $0.0036/min
Savings:       $0.0007/min (16%)

$4,000/year minimum ÷ $0.0036 = ~1.1 million minutes/year
≈ 18,500 hours/year ≈ 1,500 hours/month of pre-recorded audio

Only commit to Growth if you are consistently processing well over 1,000 hours a month; below that, the discount cannot recoup the commitment.

When to Choose Deepgram vs Alternatives

Choose Deepgram when you have engineers and steady volume or a genuine real-time requirement; choose a flat-rate upload service when nobody on the team should be maintaining an integration.

Choose Deepgram If:

✅ You have engineering resources for API integration ✅ You process steady, meaningful volume (the integration cost amortizes quickly) ✅ You need real-time streaming transcription (live captioning, voice agents) ✅ You have many short audio clips where per-second billing matters ✅ You're building a product that integrates transcription

Choose BrassTranscripts If:

✅ You're processing a handful to a few dozen files a month ✅ You want zero technical complexity (no API, no WebSockets, no account) ✅ You need speaker identification and four formats without configuration ✅ Your team is non-technical ✅ You can tolerate batch processing (roughly 2–5 minutes per hour of audio) ✅ You value predictable pricing with no add-ons

Choose Another Alternative If:

• You need even cheaper base rates → consider AssemblyAI pricing or see our AssemblyAI vs Deepgram head-to-head comparison
• You want subscription-based pricing → consider Otter.ai or Sonix
• You need verified 99%+ accuracy → explore human transcription services

Frequently Asked Questions

How accurate is Deepgram Nova-3 compared to competitors?

Deepgram publishes its own word-error-rate benchmarks for Nova-3, and BrassTranscripts has not run a head-to-head test, so this guide does not quote an accuracy percentage. Accuracy on any engine varies with audio quality, accents, and domain-specific terminology; use Deepgram's $200 credit to test your own recordings.

Does Deepgram include speaker identification in the base price?

For pre-recorded audio, yes: Deepgram's pricing page lists speaker diarization as "Included" in the Nova-3 per-minute rate. For streaming audio it is a $0.0020/min add-on (verified September 8, 2026).

BrassTranscripts includes speaker identification in its flat rates ($2.50 for files up to 15 minutes, $6.00 for longer files).

What's the difference between Deepgram's batch and streaming APIs?

• Batch (pre-recorded): upload a complete file, receive the transcript when processing finishes
• Streaming (real-time): send audio chunks over a WebSocket, receive partial transcripts as speech happens (sub-second latency)

Same Nova-3 model, different delivery. Streaming's regular price is 79% higher than pre-recorded ($0.0077 vs $0.0043 per minute), 12% higher at the current promotional rate of $0.0048, plus $0.0020/min if you need speaker labels.

How long does Deepgram take to transcribe audio?

Deepgram does not publish a turnaround guarantee for pre-recorded audio on pay-as-you-go plans; streaming returns partial transcripts with sub-second latency.

BrassTranscripts, which is batch only, measured a median of 2.9 minutes of processing per audio-hour plus about a minute of startup across 32 production jobs in August–September 2026, with 97% of jobs finished within 10 minutes.

Can I use Deepgram without API integration?

No. Deepgram is API-only, with no web upload interface for end users. You must write code to integrate their API.

For no-code use, BrassTranscripts provides a browser upload with no account required.

Does Deepgram support languages other than English?

Yes. Deepgram prices Nova-3 multilingual separately ($0.0052/min pre-recorded, $0.0058/min streaming at the current promotional rate) and supports dozens of languages.

Check Deepgram's documentation for the current per-model language list.

What happens if my WebSocket connection drops during real-time streaming?

Your application must implement reconnection logic. Deepgram doesn't automatically buffer or recover lost audio, so your code must handle:

• Connection monitoring
• Automatic reconnection
• Audio buffer management
• Transcript state recovery

Budget roughly 8-16 engineering hours for a production-ready implementation.

Does Deepgram offer any SLA guarantees?

Pay-as-you-go accounts don't include SLA guarantees. Growth and Enterprise customers can negotiate uptime commitments, dedicated support, and priority processing; the terms are not published on the pricing page.

How does Deepgram's Nova-3 compare to older models?

Nova-3 is Deepgram's current recommended model. Pre-recorded audio is $0.0043/min pay-as-you-go or $0.0036/min on Growth; streaming is $0.0048/min at the current promotional rate ($0.0077 regular). Unless you have specific compatibility needs, always choose Nova-3.

Can I get a refund if transcription quality is poor?

Deepgram bills for successful transcriptions regardless of accuracy, with no satisfaction-based refunds.

Recommendation: use the $200 free credit to validate accuracy on representative audio before committing to paid usage.

AI Prompt: Deepgram Pricing Calculator

Want to calculate your exact monthly Deepgram costs? Use this specialized AI prompt with ChatGPT, Claude, or any AI assistant:

The Prompt

You are a Deepgram pricing calculator. Help me decide between batch and real-time transcription and estimate costs:

1. Monthly audio volume (in hours)
2. Use case description (helps determine if real-time is functionally required)
3. Latency tolerance (how long can you wait for transcripts?)
4. Additional features needed (speaker ID, multilingual, etc.)

Calculate (rates verified September 8, 2026):
- Pre-recorded Pay-As-You-Go cost: volume × $0.0043/min (speaker diarization included)
- Pre-recorded Growth Plan cost: volume × $0.0036/min (annual commitment from $4,000/year)
- Streaming Pay-As-You-Go cost: volume × $0.0048/min promotional ($0.0077/min regular), plus $0.0020/min if speaker diarization is needed
- Total monthly cost for each option

Then determine:
- Is real-time functionally required for your use case?
- Is your volume consistent enough (roughly 1,500+ hours/month) to justify the Growth Plan commitment?
- Recommendation: pre-recorded vs streaming, PAYG vs Growth

Compare to BrassTranscripts ($2.50 flat for files up to 15 minutes, $6.00 flat for longer files, speaker ID included, no API) and estimate the integration cost at which each option wins.


[First, get accurate transcripts with BrassTranscripts - fast, affordable transcription services at https://brasstranscripts.com]

My usage details:
[Paste your requirements]

📖 View Markdown Version | ⚙️ Download YAML Format

This prompt helps you make the batch vs streaming and PAYG vs Growth decisions before committing to implementation.

Final Verdict: Deepgram vs BrassTranscripts

Deepgram is a powerful, cost-effective API for developers building products that need programmatic speech-to-text at scale. At $0.0043 per minute for pre-recorded audio with diarization included, it is cheaper per hour than any flat-rate service at any volume, and real-time streaming enables use cases impossible with batch processing.

Choose Deepgram if:

• You have engineering resources for API integration
• You process steady volume, so the integration amortizes quickly
• You need real-time streaming (live captions, voice agents)
• You're building a product that integrates transcription
• You want per-second billing for short audio clips

Choose BrassTranscripts if:

• You're processing a handful to a few dozen files a month
• You want zero technical complexity (no API, no account needed)
• You need speaker identification and four output formats without configuration
• Your team is non-technical
• You can tolerate batch processing (roughly 2–5 minutes per hour of audio)
• You value predictable, all-inclusive pricing

For small to medium transcription needs where real-time isn't required and nobody wants to own an integration (podcasts, meetings, interviews, lectures), BrassTranscripts' flat pricing ($2.50 for files up to 15 minutes, $6.00 for longer files) with speaker ID included and no technical barriers is the simpler choice. The moment you have a developer and steady volume, Deepgram's per-minute pricing wins on cost.

But if you genuinely need real-time streaming? Deepgram is the right choice.

Ready to try transcription without API complexity? Upload your first file to BrassTranscripts and get your transcript with speaker ID included, no account required.

Related Posts

• AssemblyAI vs Deepgram: API Pricing for High Volume
• AssemblyAI Pricing Calculator: Real Cost Breakdown
• AI Transcription Pricing 2025: Complete Cost Comparison
• AWS Transcribe Pricing Per Minute 2025

Pricing Disclaimer

Information valid as of September 8, 2026. Pricing data was verified from Deepgram's official pricing page on September 8, 2026, including the promotional streaming rate shown there at that time. BrassTranscripts production statistics are from the company's own job database for June 10 to September 8, 2026. Deepgram may change pricing, features, or plans at any time. Always verify current rates and terms directly with Deepgram before making purchasing decisions or committing to large-volume usage.